(1) A cholinergic contribution to one-trial visual recognition in the monkey was first demonstrated in studies showing that this function could be enhanced and impaired, respectively, by systemic administration of the cholinergic agonist, physostigmine, and the cholinergic antagonist, scopolamine. Later, when the entorhinal/perirhinal, or rhinal, cortex was found to be a critical substrate for recognition memory evidence was obtained that this cortex was also a critical site for the cholinergic contribution to such memory, based on the demonstration that (i) visual recognition performance was accompanied by efflux of acetylcholine in the rhinal cortex and (ii) this performance was impaired by microinfusing scopolamine directly into rhinal cortex. However, there has not been a convincing demonstration that the formation of new visual memories can be disrupted by eliminating the cholinergic input to rhinal cortex. Attempts to achieve this outcome in monkeys by injecting a neurotoxin into the basal forebrain, the major source of cholinergic projections to the cerebral cortex produced at most only modest and transient impairment in visual recognition. and even this mild effect could not be attributed directly to cholinergic denervation of temporal lobe tissue. (2) To reexamine this issue in the monkey we compared the visual recognition performance of monkeys given rhinal cortex infusions of a selective cholinergic immunotoxin, ME20.4-SAP, with that of monkeys given control infusions into this same tissue. The immunotoxin, which leads to cholinergic deafferentation of the infused cortex, yielded recognition deficits of the same magnitude as those produced by excitotoxic lesions of this region, providing the most direct demonstration to date that cholinergic activation of the rhinal cortex is essential for storing the representations of new visual stimuli and thereby enabling their later recognition. These results following cholinergic deafferentation complement our findings of those induced by blockade of perirhinal muscarinic receptors. By contrast, recognition memory is unaffected by either systemic or perirhinal injections of dopaminergic receptor antagonists (e.g. haloperidol). (3) We have also demonstrated that, like systemic injections of an N-methyl-D-asparate (NMDA) receptor antagonist (MK-801), perirhinal infusions of such an antagonist (D-AP5) impairs recognition memory. Again by contrast, recognition memory was unaffected by perirhinal injections of a kainate/AMPA receptor antagonist (CNQX). These results provide preliminary support not only for the hypothesis that stimulus memory depends on the interaction between muscarinic and NMDA receptor activation, but also for the notion that such interaction occurs within the neurons of the perirhinal cortex. Current experiments, involving perirhinal co-administration of muscarinic and NMDA receptor ligands, as well as selective immunolesioning of cholinergic afferents to rhinal cortex, will serve to refine our understanding of this interaction. (4) Our previous findings in monkeys suggested that systemic injection of haloperidol, but not of scopolamine, retards the learning of a set of concurrent visual discriminations in which the stimulus pairs within the set are each presented just once every 24 hours. In a new study, using a version of this task in which the stimulus pairs of the set are each repeated a few times within each session, systemic injections of both drugs was found to retard learning. If confirmed, the differential results on the two versions of the task would support the notion that discrimination learning with pair-repetition just once every 24 hours can be mediated only by a dopaminergic-dependent corticostriatal habit system (and, hence, is susceptible to disruption only by haloperidol), whereas learning with pair-repetition within a session is mediated by both the latter system and a cholinergic-deependent cortico-limbic memory system (and, consequently, is susceptible to disruption by both pharmacological agents). (5) The circuitry underlying the formation of stimulus memories is thought to involve a series of projections from the high-order sensory processing areas through structures in the medial temporal lobe, from there to the anterior group of thalamic nuclei and the magnocellular division of the medial dorsal nucleus (MDmc), and then to the ventral prefrontal and cingulate cortices. The parallel circuit underlying habit formation is thought to involve a series of projections from the neocortex through the basal ganglia, from there to thalamic nuclei within the ventral and intralaminar groups, and then to the premotor and supplementary motor areas. However, in the course of investigating medial thalamic efferents in macaques, we uncovered other thalamo-cortical routes that could contribute to stimulus memory and habit formation. Medial thalamic injection sites for anterograde tracers covered the midline nuclei, as well as MDmc, medial portions of the magnocellular ventral anterior nucleus (VAmc) and the intralaminar paracentral nucleus (Pc). These injections yielded terminal labeling in the outer half of layer I across an extremely large cortical expanse, sparing only the premotor and supplementary motor areas, precentral and postcentral gyri, and primary auditory cortex (the primary visual area in the occipital pole was not examined). In complementary studies, in which retrograde tracers were injected into various cortical areas, we searched for groups of neurons within the above medial thalamic region that were consistently labeled by the different injections and were therefore a potential source of the widespread projection to cortical layer I. Numerous retrogradely labeled neurons were seen in the midline group of thalamic nuclei after prefrontal, cingulate, and rhinal injections, suggesting that this particular thalamo-cortical projection could participate in the acquisition of stimulus memories. In addition, Pc and the medial portion of VAmc contained labeled cells from all the injected fields except rhinal cortex, suggesting that the widespread thalamo-cortical projections from these two nuclei, which belong to the ventral and intralaminar groups, might participate in habit formation.